Monitoring coastal northern cod: towards an optimal survey of Smith Sound, Newfoundland

• Published in: ICES Journal of Marine Science
• Main Author: Rose, George A
• Format: Article in Journal/Newspaper
• Language: English
• Published: Oxford University Press (OUP) 2003
• Subjects: Ecology; Aquatic Science; Ecology, Evolution, Behavior and Systematics; Oceanography; Gadus morhua; Newfoundland; Smith sound
• Online Access: http://dx.doi.org/10.1016/s1054-3139(03)00044-4; http://academic.oup.com/icesjms/article-pdf/60/3/453/29118895/60-3-453.pdf

Abstract The extant, coastal northern cod (Gadus morhua) have over-wintered and spawned in Smith Sound, Newfoundland, since 1995, and acoustic surveys have been conducted in several seasons since then. Cod move into the Sound in late fall, over-winter in a dense, size- and age-structured aggregation, spawn between late March and early June and then disperse into and beyond Trinity Bay during summer to feed. The optimal survey time for biomass estimation is January–February, when the waters are ice-free and the cod are in mono-specific, relatively stationary, and well-defined aggregations with the highest densities and are typically clear of “bottom returns”. Biomass surveys have been conducted in mid-January since 1999. An error analysis indicated the main sources of uncertainty to be density variability and target strength (TS). Repeated quasi-synoptic (10 h) surveys were the optimal means of producing an estimate of uncertainty about population size. Some vertical movement led to night-time surveys consistently having higher estimates than day-time surveys by approximately 15%. Detectability ranged from 73 to 86% and deadzone-corrected, acoustic measures did not differ from swept-area densities found by bottom trawling. Biomass scaling by TS used length-dependant dB/kg to reduce the size-sampling error. Overall, population biomass doubled in approximately 7 years, consistent with a rate of increase around 0.2, largely through recruitment. The surveys are internally consistent and indicate instantaneous rates of mortality among year classes of 0.3 to >2 (very high on older fish) and provide a method for monitoring the annual biomass (cv<40%).